To facilitate access to the best contextual data, Sociometrics plans to develop a Contextual Data Archive. The archive will comprise three kinds of machine-readable materials: (1) the principal data sets with the potential for use M sources of contextual data; (2) research materials - a collection of key publications and documents relating to the use of contextual data; and (3) software to find and retrieve those portions of the data sets and research materials relevant to a user's needs. The archive will offer all its products in a variety of machine-readable media, including CD-ROM. Data Sets. The heart of the archive will be approximately 20 data sets that contain contextual data describing geographic areas from the size of the block to the state. Each data set in the collection will be archived according to Sociometrics standardized conventions that have been designed to make the data sets M accessible and easy to use for secondary analysis as possible, thus encouraging their use. Each data set will include a machine-readable raw data file; machine-readable text files providing the sets programs for creating SPSS and SAS gem files for several operating platforms; and a machine-readable as well as printed User's Guide describing the content and format of the data. Research Materials. The archive will also contain a collection of publications and documents on the use of contextual data, available in both machine-readable and printed formats. The collection will emphasize two kinds of materials: research reports of analyses using contextual data in large demographic data sets; and methodological papers that address the methods and limitations of contextual data analysis. Emphasis will be given to materials that are exemplary in their use of contextual data or in there treatment of key methodological issues in contextual analysis. Software. The archive will make available search and retrieval software allowing the user to: (1) identify variables across the entire data archive meeting user-specified criteria of topic, type, and keyword, and then (2) create analysis sub-files that contain only tide desired variables. The software will, in addition, allow (3) full-text keyword search of user's guides, measurement instruments, and research materials. CD.ROM. At the end of the project the entire collection of data sets, research materials, and search and retrieval software will be placed on a single CD-ROM. PROPOSED COMMERCIAL APPLICATION: This project will produce many related commercial products. Data, documentation, and text will be sold in a variety of machine-readable formats, together with search and retrieval software. A CD-ROM containing all of the products will be made available. Experience with our other data archives leads us to expect to sell several hundred printed and machine-readable products a year. Researchers are the expected to be the primary purchasers of these products. Policy makers and practitioners are also likely purchasers. The potential market will be enhanced by the broad substantive scope of research that has the potential to make use of contextual variables, and the new value-added features.